Lecture 05 - Data Link Layer (Stop Wait Sliding Windows Ethernet FDDI)

User Server
watching with video

video clip clips
Data Link Layer Application Layer Application Layer
Beispielbild
Presentation Layer Presentation Layer
Session Layer Session Layer
Transport Layer Transport Layer
Network Layer Network Layer Network Layer
Data Link Layer Data Link Layer Data Link Layer
Physical Layer Physical Layer Physical Layer

Contents
● Design Issues
● Error Detection and Correction
● Elementary Data Link Protocols
● Sliding Window Protocols
● High Level Data Link Control (HDLC)
● Point-to-Point Protocol (PPP)
● Protocol Verification
Univ.-Prof. Dr.-Ing. Jochen H. Schiller ▪ cst.mi.fu-berlin.de ▪ Telematics ▪ Chapter 4: Data Link Layer 4.2
Design Issues
Design Issues
● The data link layer has a number
of specific functions, for this it OSI Reference Model
● provides a well-defined service Application Layer
interface to the network layer
● deals with transmission errors Presentation Layer
● regulates the flow of data Session Layer

● access to the medium
● prevent a slow receiver being not Transport Layer
swamped by a fast sender
Network Layer
Data Link Layer
Physical Layer
Layer 2: Division into two Parts
● Logical Link Control (LLC) (Layer 2b)
● Organization of the data to be sent into frames
● Guarantee (if possible) an error free transmission between neighboring nodes
through:
● Detection (and recovery) of transfer errors
● Flow Control (avoidance of overloading the receiver)
● Buffer Management
● Medium Access Control (MAC) (Layer 2a)

● Control of the access to the communication channel in broadcast networks
LLC IEEE 802.2 Logical Link Control ...

Sicherungs-
Data Link
Layer
ebene
ANSI ATM
802.3 802.4 802.5 802.6 X3T9.5 Forum
MAC ...
CSMA/CD Token Token ATM LAN
(Ethernet) Bus Ring DQDB FDDI Emulation
ISO/OSI Concrete Networks
Reale Netze
Services Provided to the Network Layer
● The data link layer provides services to the network layer
● Principal service is transferring data from the network layer of the source to the
network layer of the destination machine
● Communication of two processes on the network layer
● Virtual data path
● Actual data path
● Service provided by the data link layer
● Unacknowledged connectionless service
● No logical connection beforehand
● Source sends independent frames
● Destination does not acknowledge frames
● Acknowledged connectionless service

● No logical connection beforehand
● Destination acknowledges frames
● Acknowledged connection-oriented service

● Logical connection beforehand
● Each sent frame is numbered
● Each sent frame is received once and in the right order
Router
Packets here
Routing process Routing process
DLL DLL DLL DLL DLL DLL

Process Process Process Process Process Process
Frames here
DLL Protocol
LLC: Frame Construction
● Organization of a message into uniform units (for simpler transmission)

● Well-defined interface to the upper layer (layer 3)
● Marking of the units:
Error check
Header Data Trailer Frame Checking Sequence (FCS)
Control information (addresses, frame numbers,…)
Mark the frame by:

● Start and end flags
● Start flag and length
● Code injuries
LLC: Frame Construction
● Marking the start and end of a frame
● Character count
● Flag bytes with byte stuffing
● Starting and ending flags with bit stuffing
● Physical layer coding violations
Byte/bit-stream
Frames Frame 1 Frame 2 Frame 3
Framing: Character Count
● Specify the number of characters in the frame
Framing: Flag Bytes with Character Stuffing
● The start and end of a frame is represented by a special flag byte.
● Problem: What happens if the flag byte occurs in the data?
● Byte stuffing/character stuffing
● A special escape byte (ESC) is inserted (stuffed) by the sender and removed by
the receiver
● If the escape byte occurs in the data, then it is also stuffed
Framing: Flag Bytes with Bit Stuffing
● Character stuffing is bound to the character set (disadvantage)!
● General form: Bit stuffing
● Frames begin and end with a special pattern: 01111110
● Sender inserts after five 1s a 0-bit, i.e., 011111x...  0111110x... and the
receiver removes it
Original data:
Transmitted data:
After destuffing:
Framing: Physical Layer Coding Violations
● Only applicable if physical layer coding has some redundancy
● 1 bit data is encoded in 2 physical bits
● 1-bit coded as high-low, 0-bit coded as low-high, high-high and low-low are not
used
1-bit 0-bit not used not used
Error Detection and Correction
Error Detection and Correction
● Next task of the LLC layer: protected transmission of the frames to the
communication partner
● The transmission over layer 1 is not necessarily free of errors!
● Errors tend to come in bursts rather than single
● Question: how can errors be recognized and repaired?
● Error-detecting codes
● Error-correcting codes (Forward Error Correction, FEC)
● Scenario
● A frame consisting of m data bits and r redundant (check) bits
● Total length: n = m + r
● The n-bit unit with data and check bits is called codeword
Error-detecting and -correcting Codes: Parity bits
From the data, compute a short checksum and send it together with the data to the
receiver. The receiver also computes a checksum from the received data and
compares it with those of the sender.
Simplest procedure - parity bit: count the number of 1s:

Sender: 10111001 PB: 1 sent: 101110011 odd-Bit errors are detected
Receiver: 001011011 PB computed: 0 even-Bit errors are not detected
Corrections are not possible!
Variant: double parity
Improvement of the parity bit procedure by further parity bits. For this, several blocks
of bits are grouped and treated together:
Sender: 1011 1 Receiver: 1011 1
0010 1 0110 0
1100 0 1100 0
0110 0 0110 0
0011 0111
An incorrect bit can be identified and corrected by this procedure.
Check the terms even and odd parity! Parity bits often used in RAM (not networking)
Error Detection
Cyclic redundancy check (CRC)
Error Detection with Cyclic Codes
Problem: how to recognize errors in several bits, especially sequences of bit

errors?
The use of simple parity bits is not suitable. However, in data communication
(modem, telephone cables) such errors arise frequently.
Most often used: Polynomial Codes
Idea: a m-bit PDU (am-1, …, a0) is seen as a polynomial am-1xm-1 + … + a0 with the
coefficients ai “0” and “1”.
Example: 1100101 is interpreted as x6x5x4x3x2x1x0  x6 + x5 + x2 + 1
● For computations, polynomial arithmetic modulo 2 is used, i.e., addition
and subtraction without carriage. Both operations become Exclusive-OR
(XOR) operations.
● Polynomial arithmetic is done modulo 2

● Addition and subtraction are identical to exclusive OR (XOR), e.g.
10011011 00110011 11110000 01010101
+ 11001010 + 11001101 - 10100110 - 10101111
01010001 11111110 01010110 11111010
● Idea for error detection:
● Sender and receiver agree upon a generator polynomial G(x)
G(x) = grxr +gr-1xr-1 +…+ g1 x1+ g0x0
● The first (g0) and the last (gr) coefficient have to be 1
G(x) = xr +gr-1xr-1 +…+ g1 x1+ 1
● The sender interprets a data block of length m as polynomial M(x)
M(x) = am-1xm-1 +…+ a1x1+ a0x0
● The sender extends M(x), i.e., adds “redundant” bits in a way that the extended
polynomial M’(x) is divisible by G(x)
● Redundancy = remainder R(x) by division of the sequence with G(x)
● The receiver divides the received extended M(x) by G(x). If the remainder is 0,
there was no error, otherwise some error occurred
● Cyclic Redundancy Checksum (CRC)
● Note: also the parity bit can be seen as CRC, with generator polynomial x + 1!
● Algorithm for computing the checksum
● Let r be the degree of G(x).
● Append r zero bits to the low-order end, so it now contains m+r bits.
● The corresponding polynomial is xrM(x).
● Divide the bit string corresponding to G(x) into the bit string corresponding to
xrM(x).
● Subtract the remainder from the bit string corresponding to xrM(x).
● The result is the checksummed frame to be transmitted, denoted as T(x).
CRC: Example
Note: here, the “extra” positions are
preset with zeros – but many systems
Data to be transmitted: 10111001 like, e.g., Ethernet use the inverted
Generator polynomial: x4 + x +1  10011 bits as presets (what happens if only
0s to be transmitted?).
Sender: Receiver:
101110010000 : 10011 = 10100111 101110011001 : 10011 = 10100111
10011 10011
10000 10000
10011 10011
11100 11110
10011 10011
11110 11010
10011 10011
11010 10011
10011 10011
1001 = x3 + 1 = R(x) 0
CRC = 1001, sending 101110011001 Data received correctly
CRC is not perfect
Receiver: Receiver:
001010010001 : 10011 = 00101110 001110110001 : 10011 = 00111111
10011 10011
11110 11101
10011 10011
11010 11100
10011 10011
10010 11110
10011 10011
11 11010
10011
10011
10011
Error detected 0
Error not detected
CRC: Characteristics
● What kind of errors will be detected?
● Instead of T(x), erroneous bit string T(x)+E(x) is received
● Each 1 bit in E(x) corresponds to a bit that has been inverted
● If there are k 1 bits in E(x), k single-bit errors have occurred
● Receiver calculates: [ T(x) + E(x) ] / G(x)
● T(x)/G(x) = 0, thus result is E(x)/G(x)
● Errors that contain G(x) as a factor will be not detected
● Form of E(x)
● E(x) = xi
● Single-bit error, i determines the bit in error
● E(x) = xi + xj = xj (xi-j+ 1), i>j
● Double errors can be detected if G(x) does not divide xk+1, k up to i-j
● Most important
● A polynomial code with r check bits will detect all burst errors of length ≤r
CRC: Characteristics
Common 16-bit generator polynomials:
● CRC-16: G(x) = x16 + x15 + x2 +1
● CRC CCITT: G(x) = x16 + x12 + x5 +1
● Ethernet: G(x) = x32 + x26 + x23 + x22 + x16 + x12 + x11
+ x10 + x8 + x7 + x5 + x4 + x2 + x + 1
Error detection (for 16-bit generator polynomials):
● all single bit errors
● all double bit errors
● all three-bit errors
● all error samples with odd number of bit errors
● all error bursts with 16 or fewer bits
● 99.997% of all 17-bit error bursts
● 99.998% of all error bursts with length  18 bits
● Remaining error rate < 0.5×10-5 block error rate (original)
Not all errors are detected, an error could consist of adding a multiple of G(x) to M‘(x)
Implementation of CRC
Hardware
Computation of CRC: Shift Registers
● Implementation by Shift Registers as follows
● The register contains r bits
● There are up to r exclusive-or (XOR) gates
● The presence or absence of a gate corresponds to the presence or absence of a
term in the divisor polynomial
Computation of CRC: Shift Registers
Implementation by Shift Registers:

• XOR for subtraction
• AND for applying subtraction:
• first register = 0: no subtraction
• first register = 1: subtraction
Generator polynomial x4 + x +1: R R R R R

    
1 0 0 1 1
R R R  R  R
Simplified realization:
When no more input is given in the leftmost register, the other registers contain
the CRC result.
Shift Registers - Example
Data to be transmitted: 10111001

Generator polynomial: x4 + x +1
0 0 0 0 0 101110010000
 
Sent data
0 0 0 0 1 01110010000
1  
0 0
Sent data
0 0 0 1 0 1110010000
10  
0 0
Sent data
0 0 1 0 1 110010000
101  
0 0
Sent data
0 1 0 1 1 10010000
1011  
0 0
Sent data
1 0 1 1 1 0010000
10111  
1 1
Sent data
0 1 0 0 0 010000
101110  
0 0
Sent data
1 0 0 0 0 10000
1011100  
1 1
Sent data
0 0 1 1 1 0000
10111001  
0 0
Sent data
0 1 1 1 0 000
10111001  
0 0
Sent data
1 1 1 0 0 00
10111001  
1 1
Sent data
1 1 1 1 0 0
10111001  
1 1
Sent data
1 1 0 1 0
10111001  
1 1
Sent data
1 0 0 1 x
101110011  
1 1
Sent data
x 0 0 1 x
1011100110  
0 0
Sent data
x x 0 1 x
10111001100  
0 0
Sent data
x x x 1 x
101110011001  
0 0
Sent data
x x x x x
101110011001  
0 0
Sent data
CRC Implementation
Software
CRC – Implementation
● Shift register implementation of CRC-CCITT
● 16-bit CRC, r =16
● Generator polynomial G(x) = x16 + x12 + x5 +1
● R represents the content of the r-bit shift register
● G represents the r least-significant coefficients of the generator polynomial
● G = 0x1021
● i represents the input bit for each cycle
#include <stdio.h>
static unsigned int R; // Content of r-bit shift register
static unsigned int Gr1 = 0x0810; // G = least-significant coeff. of generator polynomial
// Gr1 = G right-shifted one bit position
void ResetCRC()
{
R = 0;
}
void UpdateCRC(char x)
{
int i; // Input bit for cycle i
int k; // Counter
printf("\nUpdateCRC(%02x)\n", x);
for (k=0; k<8; k++)
{
i = (x >> 7) & 1; // msb
printf(" %04x < %1x -> ", R, i);
if (R & 0x8000) // is msb of R == 1?
R = ((R ^ Gr1) << 1) + (i ^ 1); // ^ is xor
else
R = (R << 1) + i;
R &= 0x0ffff;
printf("%04x\n", R);
x <<= 1;
}
}
long CheckCRC(FILE *fp, int length)

{
int i;
ResetCRC();
for (i = 0; i < length; i++)
{
UpdateCRC(getc(fp));
}
return R;
}
main() {
FILE *fp;
int c, length = 0;
long crcSent;
if ((fp = fopen("tmp", "w")) == NULL) {
printf("Can't open tmp file. Exiting\n");
exit(-1);
}
ResetCRC();
while ((c = getchar()) != EOF) {
UpdateCRC(c);
putc(c, fp);
length++;
}
printf("CRC is %04x\n", R);
crcSent = R;
putc(((R >> 8) & 0xff), fp);
putc((R & 0xff), fp);
fclose(fp);
if ((fp = fopen("tmp", "r")) == NULL) {
printf("Can't read tmp file. Exiting\n");
exit(-2);
}
if (CheckCRC(fp, length) == 0)
printf("CRC checks.\n");
else
printf("Computed CRC doesn't match\n");
}
CRC: Error Correction
● In exceptional cases even errors can be corrected by CRCs.
● Example: ATM (Asynchronous Transfer Mode)
● Data units have fixed length: 5 byte header + 48 byte data
● The last header byte is a checksum for the header
● Generator polynomial G(x) = x8 + x2 + x +1
 It is even possible to correct a 1-bit error, because

there are 40 possible 1-bit errors
and those lead to 40 different non-zero remainders
● Correction is not possible with Ethernet: an Ethernet frame has a length

between 64 and 1512 byte.
Error Correction
Hamming Code
Error-detecting and -correcting Codes
● Hamming distance
● Number of places, in which two binary sequences differ
● Example: two codewords w1=10001001 and w2=10110001
10001001
XOR 10110001
00111000  d(w1, w2) = 3
● If two codewords have Hamming distance d, d single bit errors are required to
convert one codeword into the other
● Hamming distance of a code
● m bit data  2m possible data words, typically all used
● r bit check bits
● n bit codeword  2n possible codewords, typically not all used
● construct a list of all legal codewords
● find the two codewords with minimum Hamming distance
 this distance is the Hamming Distance of the whole code
● Error-detecting and error-correcting properties of a code depends on its
Hamming distance
● To detect d errors, a distance of d+1 is required
● To correct d errors, a distance of 2d+1 is required
● Example:
● Code with only four valid codewords
w1=0000000000
w2=0000011111
w3=1111100000
w4=1111111111
● Distance 5
● it can detect 4 bit errors
● it can correct 2 bit error
● If 0000000111 is received, the original must be 0000011111 (correct)
● If 0000000000 is changed to 0000000111, the error is not corrected properly
● Goal: A code with m data bits and r check bits, which can correct all single
bit errors
● Each of the 2m data bits has n illegal codewords with distance 1
● Systematically invert each of the n bits
● Each of the 2m data bits requires n+1 bit patterns
● Total number of bit sequences 2n
Requirement: (n+1)2m ≤ 2n
With: n = m+r
 (m + r + 1) ≤ 2r
● Given m, the lower limit on the number of check bits needed to correct single
errors
● The Hamming Code fulfills this lower limit
Hamming Code
● Goal: Use of several parity bits, each of them considering several bits
(overlapping). Errors can be identified and corrected by combining the
parity bits.
● The Hamming code is the “minimal” code of this category.
● Idea: Representation of each natural number by sum of powers of two.
● In a codeword with n bits: w = z1,…, zn

● the parity bits are placed exactly at the r positions, which are a power of two. At
the remaining m = n - r positions the data bits are placed.
● each of the r additional bits is a parity bit for all places, for which the
representation in powers of two contains the position of the additional bit.
Hamming Code
Number 1 2 4 8 16
● Representation of numbers as sum 1 X
2 X
of powers of 2 3 X X
4 X
● Example: 5 X X
11 = 1 + 2 + 8 6 X X
7 X X X
● A bit is protected by those bits 8 X
9 X X
occurring in its expansion 10 X X
11 X X X
● Bit 11 is checked by bits 1, 2, and 8 12 X X
13 X X X
● Bit 21 is checked by bits 1, 4, and 16 14 X X X
15 X X X X
16 X
17 X X
18 X X
19 X X X
20 X X
21 X X X
22 X X X
23 X X X X
24 X X
25 X X X
26 X X X
27 X X X X
28 X X X
Hamming Code
● Method
● m data bits
● r check bits 1 2 3 4 5 6 7 8 9 10 11
● n = m + r bits codeword
● The bits of the codeword are
numbered consecutively starting 1 2 3 4 5 6 7 8 9 10 11
with 1
● The bits that are powers of 2 are
check bits 1 2 3 4 5 6 7 8 9 10 11
● Bits: 1, 2, 4, 8, …
● The rest are filled up with data bits
● Bits: 3, 5, 6, 7, … 1 2 3 4 5 6 7 8 9 10 11
● Example
● m=7, r=4
Hamming Code
Check bits
ASCII-Code
1 2 3 4 5 6 7 8 9 10 11
H 1001000
A 1100001 0 0 1 1 0 0 1 0 0 0 0
M 1101101 1 0 1 1 1 0 0 1 0 0 1
Codeword
M 1101101 1 1 1 0 1 0 1 0 1 0 1
I 1101001 1 1 1 0 1 0 1 0 1 0 1
N 1101110
G 1100111
Receiver:
Parity bit 1: Data bit 3, 5, 7, 9, 11 3=1+2
 examine parity bits
Parity bit 2: Data bit 3, 6, 7, 10, 11 5=1+4
 if necessary, sum up indices of
Parity bit 4: Data bit 5, 6, 7 6=2+4 the incorrect parity bits
Parity bit 8: Data bit 9, 10, 11 7=1+2+4
 index of the incorrect bit
9=1+8
Problem with Hamming code: errors 1-bit errors can definitely be
involving several following bits are
10 = 2 + 8 identified and corrected
usually corrected wrong.
11 = 1 + 2 + 8
Hamming Code
Transmission
00110010000 error 00110000000 Receiver computes parity
bits:
11100000000
Summing up the indices 1, 2 and 4

 bit 7 is detected as false
a) Bit 4 and bit 11 inverted:
 parity bits 1, 2, 4, 8 are wrong
Weaknesses:
 bit 15 is to be corrected,
 2-bit errors are not corrected (or but does not exist
wrongly corrected!)
b) Bit 2 and bit 4 inverted
 3-bit errors are not recognized  parity bits 2, 4 wrong
 …  bit 6 is falsely recognized as
incorrect
c) Bits 1, 8, 9 inverted
Hamming Code is expensive in terms  all parity bits are correct
of required bits!  no error is recognized
Error Correction Mechanisms
Forward Error Correction (FEC)
● Use of error-correcting codes (see also RS- or BCH-codes)
● Falsified data in most cases can be corrected. Uncorrectable data are simply
dismissed
● Feedback from the receiver to the sender is not necessary
● Suitable for transmissions tolerating no transmission delays (video, audio) as
well as for coding resp. protecting data on CDs or DVDs
Automatic Repeat Request (ARQ)

● Use of error-detecting codes (CRC)
● Errors are detected, but cannot be corrected. Therefore, falsified data must be
requested again  retransmission of the same data!
● Introduction of flow control:
● number the data blocks to be sent
● acknowledgement of blocks by the receiver
● incorrectly transmitted data blocks are retransmitted
● Suitable for transmissions which do not tolerate errors (files)
Elementary Data Link Protocols
Implementation of Protocols
● General assumptions
● A wants to send a long stream of data to B
● Simplex protocol
● Reliable, connection-oriented service
● Infinite supply of data
● Wire-like delivery of packets, i.e., in the sent order
A B
● Example: Implementation of a protocol for layer 2
● First: Definition of some data types (protocol.h):
#define MAX_PKT 1024 /* determines packet size in bytes */
typedef enum {false, true} boolean; /* boolean type */
typedef unsigned int seq_nr; /* sequence or ack numbers */
typedef struct {
unsigned char data[MAX_PKT];
} packet; /* packet definition */
typedef enum {data, ack, nak} frame_kind; /* kinds of frames */
typedef struct { /* frames are transported in this layer */

frame_kind kind; /* what kind of a frame is it? */
seq_nr seq; /* sequence number */
seq_nr ack; /* acknowledgement number */
packet info; /* the network layer packet */
} frame;
kind seq ack info
void wait_for_event(event_type *event); // Wait for an event; return its type in event
void from_network_layer(packet *p); // Fetch a packet from the network layer

void to_network_layer(packet *p); // Deliver packet to the network layer
void from_physical_layer(frame *r); // Get frame from the physical layer

void to_physical_layer(frame *s); // Pass the frame to the physical layer
void start_timer(seq_nr k); // Start the clock running; enable timeout event
void stop_timer(seq_nr k); // Stop the clock; disable the timeout event
void start_ack_timer(void); // Start an auxiliary timer; enable ack_timeout

void stop_ack_timer(void); // Stop auxiliary timer; disable ack_timeout
void enable_network_layer(void); // Allow the network layer to cause a

// network_layer_ready event.
void disable_network_layer(void); // Forbid the network layer from causing a
// network_layer_ready event.
// Macro inc is expanded in-line: Increment k circularly.

#define inc(k) if (k < MAX_SEQ) k = k + 1; else k = 0
Network Layer
from_network_layer(packet *p) to_network_layer(packet *p)
enable_network_layer()
disable_network_layer() start_timer(seq_nr k)
Data Link Layer stop_timer(seq_nr k)
start_ack_timer(void)
stop_ack_timer(void)
to_physical_layer(frame *s) from_physical_layer(frame *r)
Physical Layer
Simplex Protocol
A B
● Protocol 1: Simplex Protocol
● Transmission in one direction
● Network layers are always ready
● Processing time is ignored
● Communication channel never damages frames
● Communication channel never loses frames
● No usage of sequence numbers or acknowledgements
● Implementation
● Two procedures
● sender1() and receiver1()
● Sender in an infinite loop

● Fetch data, send data
● Receiver in an infinite loop
● Get data, pass to network layer
Simplex Protocol: Implementation
/* Protocol 1 (utopia) provides for data transmission in one direction only, from
sender to receiver. The communication channel is assumed to be error free
and the receiver is assumed to be able to process all the input infinitely quickly.
Consequently, the sender just sits in a loop pumping data out onto the line as
fast as it can. */
typedef enum {frame_arrival} event_type; Step 1: Send without restrictions

#include "protocol.h“  No transmission errors
void sender1(void)  No flow control
{
frame s; /* buffer for an outbound frame */
packet buffer; /* buffer for an outbound packet */
while (true) {
from_network_layer(&buffer); /* go get something to send */
s.info = buffer; /* copy it into s for transmission */
to_physical_layer(&s); /* send it on its way */
}
}
void receiver1(void)
{
frame r;
event_type event; /* filled in by wait, but not used here */
while (true) {
wait_for_event(&event); /* only possibility is frame_arrival */
from_physical_layer(&r); /* go get the inbound frame */
to_network_layer(&r.info); /* pass the data to the network layer */
}
}
Simplex Protocol: Implementation
A B
from_network_layer()_ wait_for_event()
to_physical_layer() ___
from_network_layer()_ to_network_layer() ___
to_physical_layer() ___ wait_for_event() ___
from_network_layer()_ to_network_layer() ___
Important: Flow Control
● Scenario: Fast sender and slow receiver
● How to prevent a slow receiver swamped?
● Two approaches
● Feedback-based flow control
● Receiver sends information back to the sender giving permission to send more data
● Rate-based flow control
● Protocol limits the rate of data a sender may transmit without feedback from the
receiver
Simplex Stop-and-Wait Protocol
● Protocol 2: Simplex Stop-and-Wait Protocol
● Drop assumption that receiving network layer A B
processes infinitely quick Frame 0
● Main problem Ack
● How to prevent the sender from flooding the Frame 1
receiver with data
Ack
● Assumptions
● Communication channel is error free Frame 2
● Bidirectional channel Ack
● Solution
● Simple procedure: The sender sends a data
block and waits, until an acknowledgement
from the receiver arrives or a timeout is
reached.
● Incorrect blocks are repeated, otherwise the
next block is sent.
● Disadvantage: large waiting periods
between the transmission of single blocks.
Thus much transmission capacity is wasted.
Simplex Stop-and-Wait Protocol: Implementation
/* Protocol 2 (stop-and-wait) also provides for a one-directional flow of data from
sender to receiver. The communication channel is once again assumed to be error
free, as in protocol 1. However, this time, the receiver has only a finite buffer
capacity and a finite processing speed, so the protocol must explicitly prevent
the sender from flooding the receiver with data faster than it can be handled. */
typedef enum {frame_arrival} event_type; Step 2: Simple flow control

#include "protocol.h"  No transmission errors
void sender2(void)  Send-and-Wait as flow control
{
frame s; /* buffer for an outbound frame */
event_type event; /* frame_arrival is the only possibility */
while (true) {
from_network_layer(&buffer); /* go get something to send */
s.info = buffer; /* copy it into s for transmission */
to_physical_layer(&s); /* bye-bye little frame */
wait_for_event(&event); /* do not proceed until given the go ahead */
}
}
{
frame r, s; /* buffers for frames */
event_type event; /* frame_arrival is the only possibility */
while (true) {
wait_for_event(&event); /* only possibility is frame_arrival */
from_physical_layer(&r); /* go get the inbound frame */
to_physical_layer(&s); /* send a dummy frame to awaken sender */
}
}
void sender2(void)
{
{
frame s;
frame r, s;
packet buffer;
event_type event;
event_type event;
while (true) {
while (true) {
wait_for_event(&event);
from_network_layer(&buffer);
from_physical_layer(&r);
s.info = buffer;
to_network_layer(&r.info);
to_physical_layer(&s);
to_physical_layer(&s);
wait_for_event(&event);
}
}
}
}
A B
from_network_layer()_ wait_for_event() ___

to_physical_layer() ___ ___
wait_for_event() ___ from_physical_layer() _
to_network_layer()___
from_network_layer()_ wait_for_event()_ ___
to_physical_layer() __
wait_for_event() ___ from_physical_layer() _
to_network_layer()___
from_network_layer()_ wait_for_event()_ ___
Simplex Protocol for a Noisy Channel
● Protocol 3: Simplex Protocol for a Noisy
Channel
● Communication channel makes errors
Normal
● Frames may be damaged or lost completely case
● If frame is damaged, receiver detects errors
● Types of lost packets
● Frame is lost
Err Frame loss
● Ack is lost
Err Ack loss
Simplex Protocol for a Noisy Channel
● Requirement
● The sender needs a guarantee that a frame was 0
correctly received 0
● Acknowledgement from receiver
● Distinguish original frame from retransmissions m Err
● Solution: Put a sequence number into each frame m
● What is the minimum number of bits needed for the
sequence number?
n
● Enough to distinguish between frame m and m+1 pass to
 1 bit sequence number sufficient Err n network
layer
● Receiver expects a particular sequence number n
duplicate
● If arriving frame has correct sequence number n
accept, otherwise reject
Positive Acknowledgement with Retransmission
(PAR) or
Automatic Repeat reQuest (ARQ)
Simplex Protocol for a Noisy Channel: Implementation
● Step 3: Error handling and flow control – sender
/* Protocol 3 (par) allows unidirectional data flow over an unreliable channel. */
#define MAX_SEQ 1 /* must be 1 for protocol 3 */
typedef enum {frame_arrival, cksum_err, timeout} event_type;
#include "protocol.h"
void sender3(void)
{
seq_nr next_frame_to_send; /* seq number of next outgoing frame */
frame s; /* scratch variable */
event_type event;
next_frame_to_send = 0; /* initialize outbound sequence numbers */
from_network_layer(&buffer); /* fetch first packet */
while (true) {
s.info = buffer; /* construct a frame for transmission */
s.seq = next_frame_to_send; /* insert sequence number in frame */
to_physical_layer(&s); /* send it on its way */
start_timer(s.seq); /* if answer takes too long, time out */
wait_for_event(&event); /* frame_arrival, cksum_err, timeout */
if (event == frame_arrival) {
from_physical_layer(&s); /* get the acknowledgement */
if (s.ack == next_frame_to_send) {
stop_timer(s.ack); /* turn the timer off */
from_network_layer(&buffer); /* get the next one to send */
inc(next_frame_to_send); /* invert next_frame_to_send */
}
}
}
}
{
seq_nr frame_expected;
frame r, s;
event_type event;
frame_expected = 0;
while (true) {
wait_for_event(&event); /* possibilities: frame_arrival, cksum_err */
if (event == frame_arrival) { /* a valid frame has arrived. */
from_physical_layer(&r); /* go get the newly arrived frame */
if (r.seq == frame_expected) { /* this is what we have been waiting for. */
inc(frame_expected); /* next time expect the other sequence nr */
}
s.ack = 1 - frame_expected; /* tell which frame is being acked */
to_physical_layer(&s); /* send acknowledgement */
}
}
}
void sender3(void) void receiver3(void)

{ {
seq_nr next_frame_to_send; seq_nr frame_expected;
frame s; frame r, s;
packet buffer; event_type event;
event_type event; frame_expected = 0;
next_frame_to_send = 0;
from_network_layer(&buffer);
while (true) { while (true) {

s.info = buffer; wait_for_event(&event);
s.seq = next_frame_to_send; if (event == frame_arrival) {
to_physical_layer(&s); from_physical_layer(&r);
start_timer(s.seq); if (r.seq == frame_expected) {
wait_for_event(&event); to_network_layer(&r.info);
if (event == frame_arrival) { inc(frame_expected);
from_physical_layer(&s); }
if (s.ack == next_frame_to_send) { s.ack = 1 - frame_expected;
stop_timer(s.ack); to_physical_layer(&s);
from_network_layer(&buffer); }
inc(next_frame_to_send); }
} }
}
}
}
Sliding Window
● Previous protocols work only in one direction (simplex)
● What to do if full-duplex communication is required?
● Use two different simplex communication channels
● Wasting of resources, because the acknowledgements are rare and small
● Better idea: use same channel for both directions
● Data frames and acks are intermixed
● Kind-field in header distinguishes data- and ack-frames
● Piggybacking
● Instead of using special ack-packets, use a field in the header of a data frame to
inform the receiver
● When a data packet arrives, the receiver does not send immediately an ack, instead
waits a particular time interval for a data packet to the other direction
● Question: How long to wait?
● Estimate/Guess
● Fix time
● RTT
Sliding Window: Piggybacking
Data- and ack-frames Ack-frames

as individual messages piggybacked
A B A B
data A data A data
wait
ack A data B,
data ack A
data B
wait
ack B
data A,
data A ack B
ack A
Sliding Window
● Introduction of a transmission window
● Common procedure to avoid long waiting periods of the sender
● Sender and receiver agree upon a transmission window
● If W is the window size: the sender may send up to W messages without an
acknowledgement of the receiver
● Sender and receiver transmission window do not need to have the same limits
● The messages are sequentially numbered in the frame header
● MODULUS: m = 2n
● Sequence number in range of {0, 1, 2, …, 2n-1}
● In stop-and-wait n=1  Sequence number  {0, 1}
● The receiver confirms the reception of a frame by an acknowledgement (ACK)
● The sender moves the window as soon as an ACK arrives
● All frames in the window must be buffered for retransmission
● Window size n  Buffer for n frames required
● Receiver window corresponds to frames it may accept
● Frames outside the window are discarded
Sliding Window
Example: (for 3-bit sequence/acknowledgement number)
● with 3 bits for sequence/acknowledgement number, m = 23 = 8
● Stations agree upon a window size W with 1  W < m, e.g., W = 7
● The window limits the number of unacknowledged frames allowed at one time
● In this example max. 7, because of W = 7
● With receipt of an acknowledgement, the window is shifted accordingly
● Frames are numbered sequentially modulo m (for m=8, numbers from 0 to 7)
7 0 7 0 7 0 7 0 7 0
ACK
6 0 ACK1 2 6 ACK1 6 6 1 6 1 6 1
5 2 5 2 5 2 5 2 5 2
0 41 3 2 3 4 453 6 7 0 4 13 2 3 4 4 53 6 7 04 13 2
Station
Station sendet
sends Stationreceives
Station erhält Station verschiebt
Station slides Stationreceives
Station erhält Station verschiebt
Station slides
Frames
frames 00- -66 Quittung
ack 0 0 window byum
Fenster 1,
1 and Quittung
ack 1,21, 2 window byum
Fenster 2,
2 and
sendet Frame7 7
sends frame sendet Frame 0,
sends frames 0,1
1
Time
Maximum Window Size with Sliding Window
● There is a reason why window size W has to be smaller than MODULUS 2n
● In the example, sequence numbers have 3 bits
 sequence numbers {0, …, 7}
 MODULUS m = 8
● Assume the window size to be W=8 and A sends 3 frames to B
● B acknowledges frame 2 (ACK 3 was sent to A)
● B has acknowledged 0 1 2
● A sends 8 frames, without receiving an acknowledgement from B
● A sends 0 1 2 3 4 5 6 7 0 1 2
● A receives an acknowledgement ACK 3.
● Case 1: B only has received 0 1 2
● Case 2: B has received 0 1 2 3 4 5 6 7 0 1 2
● A does not know whether case 1 or 2 holds for B, the acknowledgement is
not clear!
 W  7 (W < MODULUS in general)
Sliding Window
● Protocol 4: A one-bit sliding window protocol
● Special case with max. window size of 1  Stop-and-wait
● Sender transmits a frame and waits
A One-bit Sliding Window Protocol: Implementation
/* Protocol 4 (sliding window) is bidirectional. */
#define MAX_SEQ 1 /* must be 1 for protocol 4 */
typedef enum {frame_arrival, cksum_err, timeout} event_type;
void protocol4 (void)

{
seq_nr next_frame_to_send; /* 0 or 1 only */
seq_nr frame_expected; /* 0 or 1 only */
frame r, s; /* scratch variables */
packet buffer; /* current packet being sent */
event_type event;
next_frame_to_send = 0; /* next frame on the outbound stream */

frame_expected = 0; /* frame expected next */
from_network_layer(&buffer); /* fetch a packet from the network layer */
s.info = buffer; /* prepare to send the initial frame */
s.seq = next_frame_to_send; /* insert sequence number into frame */
s.ack = 1 - frame_expected; /* piggybacked ack */
to_physical_layer(&s); /* transmit the frame */
start_timer(s.seq); /* start the timer running */
// while() part – see next slide
}
Only one side
has to run that!
while (true) {
wait_for_event(&event); /* frame_arrival, cksum_err, or timeout */
if (event == frame_arrival) { /* a frame has arrived undamaged. */
from_physical_layer(&r); /* go get it */
if (r.seq == frame_expected) { /* handle inbound frame stream. */

to_network_layer(&r.info); /* pass packet to network layer */
inc(frame_expected); /* invert seq number expected next */
}
if (r.ack == next_frame_to_send) { /* handle outbound frame stream. */

stop_timer(r.ack); /* turn the timer off */
from_network_layer(&buffer); /* fetch new pkt from network layer */
inc(next_frame_to_send); /* invert sender's sequence number */
}
}
s.info = buffer; /* construct outbound frame */
s.seq = next_frame_to_send; /* insert sequence number into it */
s.ack = 1 - frame_expected; /* seq number of last received frame */
to_physical_layer(&s); /* transmit a frame */
start_timer(s.seq); /* start the timer running */
}
Sender Receiver
while (true) { while (true) {
wait_for_event(&event); wait_for_event(&event);
if (event == frame_arrival) { if (event == frame_arrival) {
from_physical_layer(&r); from_physical_layer(&r);
if (r.seq == frame_expected) { if (r.seq == frame_expected) {

to_network_layer(&r.info); to_network_layer(&r.info);
inc(frame_expected); inc(frame_expected);
} }
if (r.ack == next_frame_to_send) { if (r.ack == next_frame_to_send) {

stop_timer(r.ack); stop_timer(r.ack);
from_network_layer(&buffer); from_network_layer(&buffer);
inc(next_frame_to_send); inc(next_frame_to_send);
} }
} }
s.info = buffer; s.info = buffer;
s.seq = next_frame_to_send; s.seq = next_frame_to_send;
s.ack = 1 - frame_expected; s.ack = 1 - frame_expected;
to_physical_layer(&s); to_physical_layer(&s);
start_timer(s.seq); start_timer(s.seq);
} }
A One-bit Sliding Window Protocol
Normal operation Particular situation when both
of the protocol sides send an initial frame
A B A B
nf=0 nf=0 nf=0 nf=0
ex=0 ex=0 ex=0 0,1,A0 ex=0
0,1,A0
0,1,B0
ex=1 ex=1
0,0,B0 ex=1 0,0,B0
ex=1 nf=1 0,0,A0

nf=1
1,0,A1 1,0,A1
ex=2=0
nf=1 nf = next frame
ex = expected frame
1,1,B1 Packet format is
(seq, ack, packet number)
Half of the frames contain

duplicates, even without
transmission errors!
Sliding Window
● Until now transmission times assumed
negligible
t=0
● Example scenario
● Long round-trip time 20 ms frame
● 50 kbps satellite channel with 500 msec

round-trip propagation time 270 ms
● Transmission of 1000-bit frame
● Sender is blocked 500/520=96% of the
time ack
● Utilization only 4%
520 ms
time
Sliding Window
● Source of problem ● Pipelining
● Sender has to wait for an ack ● Assumptions
before sending next frame ● Channel capacity is b bits/sec
● Solution ● Frame size l bits
● Allow sender to transmit up to W ● Round-trip-time R sec
frames before blocking ● Time to transmit a frame: l/b sec
● In the example W=520/20=26 ● After sending last bit, delay of R/2 sec
 Maximum window size = 26 ● Another R/2 sec delay for ack
● If bandwidth × round-trip-delay large, ● Example: Stop-and-wait
large window is required
● Busy for l/b sec
 Capacity of the pipe
● Idle for R sec
● Sender has to fill the pipe
● Utilization l/(l+bR)
● If l<bR  Utilization < ½
Pipelining
● If l=bR  Utilization = ½
Sliding Window: Pipelining and Go-back-N
● Go-back-N
● Sender transmits frames according the window size
● When timer expires for a frame, sender retransmits all buffered frames
● Receiver:
● Correct frames are acked
● Incorrect frames are discarded and no ack is send
Timeout for 2
0 1 2 3 4 5 2 3 4 5 6
Source
Destination
ACK0 ACK1 ACK2 ACK3
Discard
Sliding Window:
Pipelining and Go-back-N: Implementation
/* Protocol 5 (go back n) allows multiple outstanding frames. The sender may transmit up
to MAX_SEQ frames without waiting for an ack. In addition, unlike in the previous
protocols, the network layer is not assumed to have a new packet all the time. Instead,
the network layer causes a network_layer_ready event when there is a packet to send. */
#define MAX_SEQ 7 /* should be 2^n - 1 */

typedef enum {frame_arrival, cksum_err, timeout, network_layer_ready} event_type;
static boolean between(seq_nr a, seq_nr b, seq_nr c)

{
/* Return true if a <= b < c circularly; false otherwise. */
if (((a <= b) && (b < c)) || ((c < a) && (a <= b)) || ((b < c) && (c < a)))
return(true);
else
return(false);
}
static void send_data(seq_nr frame_nr, seq_nr frame_expected, packet buffer[])

{
/* Construct and send a data frame. */
s.info = buffer[frame_nr]; /* insert packet into frame */
s.seq = frame_nr; /* insert sequence number into frame */
s.ack = (frame_expected + MAX_SEQ)%(MAX_SEQ + 1); /* piggyback ack */
to_physical_layer(&s); /* transmit the frame */
start_timer(frame_nr); /* start the timer running */
}
Sliding Window:
void protocol5(void)
{
seq_nr next_frame_to_send; /* MAX_SEQ > 1; used for outbound stream */
seq_nr ack_expected; /* oldest frame as yet unacknowledged */
seq_nr frame_expected; /* next frame expected on inbound stream */
frame r; /* scratch variable */
packet buffer[MAX_SEQ + 1]; /* buffers for the outbound stream */
seq_nr nbuffered; /* # output buffers currently in use */
seq_nr i; /* used to index into the buffer array */
event_type event;
enable_network_layer(); /* allow network_layer_ready events */

ack_expected = 0; /* next ack expected inbound */
next_frame_to_send = 0; /* next frame going out */
frame_expected = 0; /* number of frame expected inbound */
nbuffered = 0; /* initially no packets are buffered */
Sliding Window:
while (true) {
wait_for_event(&event); /* four possibilities: see event_type above*/
switch(event) {
case network_layer_ready: /* the network layer has a packet to send */
/* Accept, save, and transmit a new frame. */
from_network_layer(&buffer[next_frame_to_send]); /* fetch new packet */
nbuffered = nbuffered + 1; /* expand the sender's window */
send_data(next_frame_to_send,frame_expected, buffer); /* transmit the frame */
inc(next_frame_to_send); /* advance sender's upper window edge */
break;
case frame_arrival: /* a data or control frame has arrived */

from_physical_layer(&r); /* get incoming frame from physical layer */
if (r.seq == frame_expected) { /* Frames are accepted only in order. */

to_network_layer(&r.info); /* pass packet to network layer */
inc(frame_expected); /* advance lower edge of recv's window */
}
/* Ack n implies n-1, n-2, etc. */
/* Check for this. */
while (between(ack_expected, r.ack, next_frame_to_send)) {
/* Handle piggybacked ack. */
nbuffered = nbuffered - 1; /* one frame fewer buffered */
stop_timer(ack_expected); /* frame arrived intact; stop timer */
inc(ack_expected); /* contract sender's window */
}
break;
Sliding Window:
case cksum_err: break; /* just ignore bad frames */
case timeout: /* trouble; retransmit all outstanding frames */

next_frame_to_send = ack_expected; /* start retransmitting here */
for (i = 1; i <= nbuffered; i++) {
send_data(next_frame_to_send, frame_expected, buffer); /* resend 1 frame */
inc(next_frame_to_send); /* prepare to send the next one */
}
}// of switch
if (nbuffered < MAX_SEQ)

enable_network_layer();
else
disable_network_layer();
}// of while
}
Sliding Window: Pipelining and Selective Repeat
● Selective Repeat (SREPEAT)
● Receiver
● When a frame is received correct, send ack
● When a frame is missing buffer following correct frames
● When the missing frame arrives, send an ack for the subsequently received frames
● Sender
● Send frames according sliding window
● If an ack does not arrive before timeout, repeat all frames (like in Go-back-N)
● When the sender receives the ack, it stops repeating old frames and goes on with new
frames
Thus, the capacity is used more efficiently, but the receiver needs more
buffer
Sliding Window:
Pipelining and Selective Repeat: Implementation
/* Protocol 6 (selective repeat) accepts frames out of order but passes packets to the
network layer in order. Associated with each outstanding frame is a timer. When the timer
expires, only that frame is retransmitted, not all the outstanding frames, as in protocol 5. */
#define MAX_SEQ 7 /* should be 2^n - 1 */

#define NR_BUFS ((MAX_SEQ + 1)/2)
typedef enum { frame_arrival, cksum_err, timeout, network_layer_ready, ack_timeout } event_type;
boolean no_nak = true; /* no nak has been sent yet */
seq_nr oldest_frame = MAX_SEQ + 1; /* initial value is only for the simulator */
static boolean between (seq_nr a, seq_nr b, seq_nr c)

{/* Same as between in protocol5, but shorter and more obscure. */
return ((a <= b) && (b < c)) || ((c < a) && (a <= b)) || ((b < c) && (c < a));
}
static void send_frame (frame_kind fk, seq_nr frame_nr, seq_nr frame_expected, packet buffer[])
{ /* Constr. and send a data, ack, nak frame.*/
s.kind = fk; /* kind == data, ack, or nak */
if (fk == data)
s.info = buffer[frame_nr % NR_BUFS];
s.seq = frame_nr; /* only meaningful for data frames */
s.ack = (frame_expected + MAX_SEQ) % (MAX_SEQ + 1);
if (fk == nak)
no_nak = false; /* one nak per frame, please */
to_physical_layer (&s); /* transmit the frame */
if (fk == data)
start_timer (frame_nr % NR_BUFS);
stop_ack_timer (); /* no need for separate ack frame */
}
Sliding Window:
void protocol6 (void)
{
seq_nr ack_expected; /* lower edge of sender's window */
seq_nr next_frame_to_send; /* upper edge of sender's window + 1 */
seq_nr frame_expected; /* lower edge of receiver's window */
seq_nr too_far; /* upper edge of receiver's window + 1 */
int i; /* index into buffer pool */
frame r; /* scratch variable */
packet out_buf[NR_BUFS]; /* buffers for the outbound stream */
packet in_buf[NR_BUFS]; /* buffers for the inbound stream */
boolean arrived[NR_BUFS]; /* inbound bit map */
seq_nr nbuffered; /* how many output buffers currently used */
event_type event;
enable_network_layer (); /* initialize */

ack_expected = 0; /* next ack expected on the inbound stream */
next_frame_to_send = 0; /* number of next outgoing frame */
frame_expected = 0;
too_far = NR_BUFS;
nbuffered = 0; /* initially no packets are buffered */
for (i = 0; i < NR_BUFS; i++)

arrived[i] = false;
Sliding Window:
while (true) {
wait_for_event (&event); /* five possibilities: see event_type above */
switch (event) {
case network_layer_ready: /* accept, save, and transmit a new frame */
nbuffered = nbuffered + 1; /* expand the window */
from_network_layer (&out_buf[next_frame_to_send % NR_BUFS]); /* fetch new packet */
send_frame (data, next_frame_to_send, frame_expected, out_buf); /* transmit the frame */
inc (next_frame_to_send); /* advance upper window edge */
break;
case frame_arrival: /* a data or control frame has arrived */
from_physical_layer (&r); /* fetch incoming frame from physical layer */
if (r.kind == data){ /* An undamaged frame has arrived. */
if ((r.seq != frame_expected) && no_nak)
send_frame (nak, 0, frame_expected, out_buf);
else
start_ack_timer ();
if (between (frame_expected, r.seq, too_far) && (arrived[r.seq % NR_BUFS] == false))
{ /* Frames may be accepted in any order. */
arrived[r.seq % NR_BUFS] = true; /* mark buffer as full */
in_buf[r.seq % NR_BUFS] = r.info; /* insert data into buffer */
while (arrived[frame_expected % NR_BUFS]) {
/* Pass frames and advance window. */
to_network_layer (&in_buf[frame_expected % NR_BUFS]);
no_nak = true;
arrived[frame_expected % NR_BUFS] = false;
inc (frame_expected); /* advance lower edge of receiver's window */
inc (too_far); /* advance upper edge of receiver's window */
start_ack_timer (); /* to see if a separate ack is needed */
}
}// if
}// switch
Sliding Window:
if ((r.kind == nak) && between (ack_expected, (r.ack + 1) % (MAX_SEQ + 1), next_frame_to_send))
send_frame (data, (r.ack + 1) % (MAX_SEQ + 1), frame_expected, out_buf);
while (between (ack_expected, r.ack, next_frame_to_send))
{
nbuffered = nbuffered - 1; /* handle piggybacked ack */
stop_timer (ack_expected % NR_BUFS); /* frame arrived intact */
inc (ack_expected); /* advance lower edge of sender's window */
}
break;
case cksum_err:
if (no_nak)
send_frame (nak, 0, frame_expected, out_buf); /* damaged frame */
break;
case timeout:
send_frame (data, oldest_frame, frame_expected, out_buf); /* we timed out */
break;
case ack_timeout:
send_frame (ack, 0, frame_expected, out_buf); /* ack timer expired; send ack */
}
if (nbuffered < NR_BUFS)

enable_network_layer ();
else
disable_network_layer ();
}
}
Sliding Window: Pipelining and Selective Reject
● Selective Reject SREJj
● Like in SREPEAT, correctly received frames after a missing frame are buffered
● The receiver sends a negative acknowledgement for the missing frame j
● The sender repeats only frame j
● By this, no unnecessary duplicates are sent, the efficiency of capacity usage again is
enhanced.
● Variant
● The receiver can send a list of missing frames to the sender, not only single negative
acknowledgements
● But: again the receiver needs more buffer.
Sliding Window: Retransmission Strategies
Go-Back-N Selective Repeat Selective Reject
Sender Receiver Sender Receiver Sender Receiver
1 1 1
2 2 2
3 3 3
Timeout for 2
Timeout for 2
4 1 4 1 4 1
Timeout for 2
5 5 5
6 6 3 6 3
Buffer frames
7 7 4 7 4
8 8 5 8 5
9 9 6 2 6
2 2 7 9 7
3 3 8 10 8
4 4 9 11 2
5 2 5 2 12 9
6 3 6 13 10
7 4 7 14 11
8 5 10 15 12
9 6 11 16 13
10 7 12 10 17 14
8 11 18 15
19 16
17
BE WARNED - two concepts mixed: error correction via ARQ and flow control!
Implementation of Timer
● Some protocols require many timers, but few hardware timers exist
● Implement timers in software by using one hardware timer
● Store expiration times in a linked list and update it during protocol runtime
High Level Data Link Control (HDLC)
Synchronous Data Link Control (SDLC) by IBM
Advanced Data Communication Control High-level Data Link Control

Procedure (ADCCP) by ANSI (HDLC) by ISO
Link Access Procedure (LAP)

by CCITT
Link Access Procedure, Balanced

Current standard for (LAPB) by CCITT
HDLC is ISO 13239
● Three types of stations
● Primary station: responsible for controlling the operation of the link.
● Frames are called commands
● Secondary station: operates under the control of the primary station.
● Frames are called responses.
● Combined station: combination of primary and secondary.
● Link configurations
● Unbalanced configuration
● One primary and one or more secondary stations
● Full-duplex and half-duplex transmissions
● Balanced configuration
● Two combined stations
● Full-duplex and half-duplex transmissions
● Data transfer modes
● Normal response mode
● Primary station initiates communication and secondary responses
● One computer controls several terminals
- Computer polls the terminals for input
● Asynchronous balanced mode

● Balanced configuration
● Either combined station initiates communication
● Most widely used one, no polling
● Asynchronous response mode
● Secondary initiates communication
● Primary controls the line
● Rarely used, for special situations
● Bit oriented protocol
● Frame identification
● Mark the beginning and the end of a frame with a flag: “01111110”
● Flag may never occur within a frame: bit stuffing
● Sender inserts a “0” after each sequence of five “1”
● The receiver removes this zero
Sender Receiver
11111111 11111111
111110111 111110111
Bits 8 8 8 ≥0 16 8
01111110 Address Control Data Checksum 01111110
For synchronization on
layer 1
● Address
● contains the address of the receiver, only important when multiple terminal on
line
● on point-to-point lines used to distinguish commands from responses
● Control
● sequence numbers, acknowledgements, and some other purposes
● Data
● arbitrary long, however efficiency of the checksum falls of with increasing frame
length
● Checksum
● is computed by using a CRC
● Minimum frame consists of 32 bit, excluding the flags
● Three kinds of frames
● Information (I-Frame)
● Information frames transport user data from the network layer
● They can include flow and error control information piggybacked on data. The
subfields in the control field define these functions
● Supervisory (S-Frame)
● Supervisory frames are used for flow and error control when piggybacking is not
possible
● Unnumbered (U-Frame)
● Unnumbered frames are used for link management
● Can carry data when unreliable connectionless service is required
● They exchange session management and control information between connected
devices
● Control field of the header 8 8 8 ≥0 16 8
distinguishes the frame types 01111110 Address Control Data Checksum 01111110
● Seq: sequence number

● HDLC uses sliding window with a 3-bit
sequence number
● Next: piggybacked acknowledgement
Bits 1 3 1 3
● Number of the first frame not received,
i.e., the next frame expected I-Frame 0 Seq P/F Next
● P/F: Poll/Final
● Used when a master polls a group of S-Frame 1 0 Type P/F Next
terminals
● If P: Terminal is invited to send data. All
frames of the terminal have the P/F bit U-Frame 1 1 Type P/F Modifier
set to P, only the last frame is set to F
● Type: used to distinguish various
kinds of frames
Fields
Type Name Send frame Seq., ACK up to Next-1 in the other
1 2 3 4 5 6 7 8
direction
I
I 0 Seq. P Next
(Data Frame) Ready to receive, ACK up to Next-1 in the other
RR P/ direction
1 0 0 0 F Next
(Receive Ready)
Temporarily not ready to receive, acknowledge up to
RNR
1 0 1 0
P/
Next Next-1 in the other direction
F
(Receive not Ready)
S
REJ P/ NACK for Next, ACK up to Next-1. The sender should
1 0 0 1 Next
(Reject) F repeat everything beginning with Next
SREJ P/
1 0 1 1 F Next ACK up to Next-1, selective NACK for Next
(Selective Reject)
SABM
(Set Asynchronous 1 1 1 1 P 1 0 0 Connection Establishment
Balanced Mode)
DISC
1 1 0 0 P 0 1 0 Announcement of connection termination
(Disconnect)
UA
1 1 0 0 F 1 1 0 General ACK (e.g. for connection establishment)
U (Unnumbered ACK)
CMDR
1 1 1 0 F 0 0 1
(Command Reject) Frame/command not valid (invalid frame, wrong
FRMR sequence number, …)
1 1 1 0 F 0 0 1
(Frame Reject)
DM
1 1 1 1 F 0 0 0 Connection Termination
(Disconnect Mode)
● There are some special commands provided
● DISC (Disconnect): machine announces that it is going down
● SNRM (Set Normal Response Time): machine announces that it is back
● Sequence numbers are set to zero
● SABM (Set Asynchronous Balanced Mode)
● Resets the line and sets both devices equal
● Enhancements
● 7-bit sequence numbers, instead of 3-bit sequence numbers
Point-to-Point Protocol (PPP)
Classical Modem access to the Internet – somewhat historical
● A home personal computer acting as an internet host.
● Point-to-point connections to the Internet
● Establish a direct connection between two nodes
● Dial-up host-router connections
● Router-router connection
● PPP is defined in RFC 1661, RFC 1662, RFC 1663
● Handles error detection, support multiple protocols
● Supports synchronous and asynchronous connections
● Features of PPP
● Framing method with error detection
● Link Control Protocol (LCP)
● Initiation of connections
● Testing of connections
● Negotiation of options
● Terminating of connections
● Network Control Protocol (NCP) for each network layer supported
● Negotiate network-layer options, e.g., network address or compression options,
after the connection has been established.
● PPP frame format was chosen closely to resemble the HDLC frame format
● Major difference is that PPP is character oriented and uses byte-stuffing
●The structure of frames is oriented at HDLC:

● Flag: same as for HDLC
● Address: unnecessary, therefore set to 11111111
● Control: unnumbered mode (marked by 00000011) without sequence numbers
and acknowledgments as default procedure
● Protocol: specifies contents of the data part (Payload), i.e., contains an
identifier to inform the receiver about what to do.
● “00000110” means “IP protocol used for processing the data”.
● Checksum: 2 byte error detection is usually used
● The Link Control Protocol (LCP) frame types defined in RFC 1661
● I = Initiator proposes option values
● R = Responder accepts or rejects proposed options
Name Direction Description

Configure-request IR List of proposed options and values
Configure-ack IR All options are accepted
Configure-nak IR Some options are not accepted
Configure-reject IR Some options are not negotiable
Terminate-request IR Request to shut the line down
Terminate-ack IR OK, line shut down
Code-reject IR Unknown request received
Protocol-reject IR Unknown protocol requested
Echo-request IR Please send this frame back
Echo-reply IR Here is the frame back
Discard-request IR Just discard this frame (for testing)
Some more PPP flavors
● PPPoE (PPP over Ethernet, RFC 2516)
● Encapsulation of PPP frames inside Ethernet frames
● Often used in the context of DSL – fits into telco architectures (virtually a
connection over a Ethernet
● Relatively high overhead with small packets
● Can be replaced by DHCP
● PPPoA (PPP over ATM, RFC 2364: PPP over AAL5)

● Encapsulation of PPP frames in AAL5
● Used in the context of DSL or cable modem
Protocol Verification
● Realistic protocols and their implementations are very complicated
● How to verify that an implementation of a protocol is correct?
● Two concepts for the verification of protocols

● Finite State Machine (FSM) Models
● Petri Net Models
Finite State Machine (FSM) Models
Finite State Machine Models
● Construct a finite state machine for a protocol  Protocol machine
● The machine is always in a specific state
● The states consist of all the values of its variables
● Often, a large number of states can be grouped
● Number of state is 2n, where n is the number of bits needed to represent all
variables
● Well-known techniques from graph theory allow the determination of which
states are reachable and which are not
● Reachability analysis
● Incompleteness: protocol machine is in a state and the protocol does not specify what
to do
● Deadlock: no exit or progress from a state
● Extraneous transition: event occurs in a state in which it should not occur
Finite State Machine Models
● Formal definition of a protocol machine as a quadruple (S, M, I, T)
● S set of states the processes and channel can be in
● M set of frames that can be exchanged over the channel
● I set of initial states of the processes
● T set of transitions between states
Finite State Machine Models: Example Protocol 3
● Example: Protocol 3
● Unidirectional data flow over an unreliable channel
● Channel has 4 possible states  2 bits
● 0 frame in transmission from sender to receiver
● 1 frame in transmission from sender to receiver
● Ack frame in transmission from receiver to sender
● Empty channel
● Sender and receiver have 2 possible states  2 bits
● The whole system has 16 possible states  4 bits
● Each state is labeled by three characters: (SRC)
● S  {0 = sender transmits a 0-frame, 1 = sender transmits a 1-frame}
● R  {0 = receiver expects 0-frame, 1 = receiver expects 1-frame}
● C  {0 = 0-frame, 1 = 1-frame, A = ack-frame, - = empty}
● Examples
● (000) Sender sends 0-frame, receiver expects 0-frame, 0-frame on channel
● (01-) Sender sends 0-frame, receiver expects 1-frame, channel is empty
Protocol 3 Protocol 3 Protocol 3
Normal operation Loss of a data frame Loss of an ack
A B A B A B
nf=0 ex=0 nf=0 ex=0 nf=0 ex=0
seq=0 seq=0 seq=0
?seq = ex ?seq = ex
ex=1 ex=1
ack=0
ack=0
?nf = ack
nf=1 Timeout Timeout
seq=1 seq=0 seq=0

?seq ≠ ex
?seq = ex ex=1
ex=0
ack=0
ack=1
?nf = ack
nf=0
State diagram for protocol 3 Transitions
Unreachable states are not shown.
Protocol 3
Normal operation
A B
nf=0 ex=0
seq=0
?seq = ex
ex=1 000 01A
ack=0
?nf = ack
nf=1
seq=1
?seq = ex
ex=0 10A 111
ack=1
?nf = ack
nf=0
ack-frame lost
0-frame lost
1-frame lost
ack-frame lost
Petri Net Models
Petri Net Models
● Petri Net ● Dynamics of Petri Nets
● A Petri Net has four basic elements ● A transition is enabled if there are
● Places at least one token in each of its
- drawn as circles input places
- represent a state of the system ● Any enabled transition may fire at
● Transitions will
- drawn as horizontal or vertical bar ● Tokens are removed from input
- have 0 or more inputs arcs places and created in output places
- have 0 or more output arcs ● If several transitions are enabled
● Arcs any of them may fire
- connect places and transitions ● The choice of the firing transition is
● Tokens indeterminate
- drawn as filled circles
- number of resources in state
Petri Net Models
● A Petri Net is defined as a tuple
PN = (P, T, F, W, M0)
● with
● Places: P = {P1, P2, …, Pm}
● Transitions: T = {T1, T2, …, Tn}
● Arcs: F(PxT)(TxP)
● Weight function: W: F  {1, 2, 3, … }
● Initial marking: M0: P  {0, 1, 2, … }
Petri Net Models: Examples
2
3
2
3
Petri Net Models
A Petri net with two places and two transitions.
Transition 1 is enabled, transition 2 is not
After transition 1 has fired, transition 2 is enabled
Petri Net Models: Example Protocol 3
A Petri net model for protocol 3.

Transition 10
Transition 11
Transition 1
(next slide)
Transition 5
Summary
● The data link layer deals with networking of adjacent nodes
● The data link layer consists of two sublayers
● Logical Link Control (LLC)
● Media Access Control (MAC)
● Its main task is to provide services to the network layer
● This includes the delivery of error free frames, which contain payload from
the network layer
● Transmission error
● Flow control
● Implementation of protocols is very complicated and weird
● Formal methods for their verification are needed
● Finite state machines
● Petri nets

Lecture 05 - Data Link Layer (Stop Wait Sliding Windows Ethernet FDDI)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture 05 - Data Link Layer (Stop Wait Sliding Windows Ethernet FDDI)

Uploaded by

Copyright:

Available Formats

User Server

watching with video

Data Link Layer Application Layer Application Layer

Session Layer Session Layer

Transport Layer Transport Layer

Network Layer Network Layer Network Layer

Data Link Layer Data Link Layer Data Link Layer

Physical Layer Physical Layer Physical Layer

● regulates the flow of data Session Layer

Data Link Layer

● Medium Access Control (MAC) (Layer 2a)

LLC IEEE 802.2 Logical Link Control ...

● Acknowledged connectionless service

● Acknowledged connection-oriented service

DLL DLL DLL DLL DLL DLL

● Organization of a message into uniform units (for simpler transmission)

Control information (addresses, frame numbers,…)

Mark the frame by:

1-bit 0-bit not used not used

Simplest procedure - parity bit: count the number of 1s:

Problem: how to recognize errors in several bits, especially sequences of bit

Most often used: Polynomial Codes

Example: 1100101 is interpreted as x6x5x4x3x2x1x0  x6 + x5 + x2 + 1

● Polynomial arithmetic is done modulo 2

CRC = 1001, sending 101110011001 Data received correctly

Error not detected

Implementation by Shift Registers:

Generator polynomial x4 + x +1: R R R R R

Data to be transmitted: 10111001

long CheckCRC(FILE *fp, int length)

 It is even possible to correct a 1-bit error, because

● Correction is not possible with Ethernet: an Ethernet frame has a length

● Idea: Representation of each natural number by sum of powers of two.

● In a codeword with n bits: w = z1,…, zn

Summing up the indices 1, 2 and 4

Automatic Repeat Request (ARQ)

typedef enum {data, ack, nak} frame_kind; /* kinds of frames */

typedef struct { /* frames are transported in this layer */

kind seq ack info

void from_network_layer(packet *p); // Fetch a packet from the network layer

void from_physical_layer(frame *r); // Get frame from the physical layer

void start_ack_timer(void); // Start an auxiliary timer; enable ack_timeout

void enable_network_layer(void); // Allow the network layer to cause a

// Macro inc is expanded in-line: Increment k circularly.

from_network_layer(packet *p) to_network_layer(packet *p)

to_physical_layer(frame *s) from_physical_layer(frame *r)

● Sender in an infinite loop

typedef enum {frame_arrival} event_type; Step 1: Send without restrictions

typedef enum {frame_arrival} event_type; Step 2: Simple flow control

from_network_layer()_ wait_for_event() ___

Err Ack loss

void sender3(void) void receiver3(void)

while (true) { while (true) {

Data- and ack-frames Ack-frames

#define MAX_SEQ 1 /* must be 1 for protocol 4 */

typedef enum {frame_arrival, cksum_err, timeout} event_type;

void protocol4 (void)

next_frame_to_send = 0; /* next frame on the outbound stream */

// while() part – see next slide

if (r.seq == frame_expected) { /* handle inbound frame stream. */

if (r.ack == next_frame_to_send) { /* handle outbound frame stream. */

if (r.seq == frame_expected) { if (r.seq == frame_expected) {

if (r.ack == next_frame_to_send) { if (r.ack == next_frame_to_send) {

0,0,B0 ex=1 0,0,B0

from_network_layer(packet p) to_network_layer(packet p)

to_physical_layer(frame s) from_physical_layer(frame r)